Infection Risk Prediction in Healthcare Settings: Lessons from COVID-19 Contact Tracing

Area of Science:

• Epidemiology
• Infectious Disease Control
• Public Health

Background:

• Contact tracing is crucial for managing infectious disease outbreaks, including SARS-CoV-2.
• Effective contact tracing strategies are needed to predict infection risk among healthcare workers (HCWs) without overwhelming systems.

Purpose of the Study:

• To evaluate the predictive performance of contact-based and risk assessment models for SARS-CoV-2 infection in exposed HCWs.
• To determine if incorporating symptom status improves these predictive models.

Main Methods:

• Retrospective cohort study of 3389 exposed HCWs at a tertiary university hospital.
• Development of logistic regression models based on contact tracing data (mask use, duration, distance) and risk assessment (low, medium, high).
• Comparison of model performance with and without inclusion of symptom data, adjusting for covariates.

Main Results:

• RT-PCR positivity was 5.9% among tested HCWs.
• Lack of PPE and longer contact duration (≥15 min) were associated with increased SARS-CoV-2 positivity.
• High-risk contact status significantly increased infection odds (OR=2.76).
• Inclusion of symptom data improved model predictive parameters (AIC, Tjur's R²).

Conclusions:

• Symptom assessment enhances the accuracy of contact-based and risk assessment models for predicting SARS-CoV-2 infection in HCWs.
• Institutions should integrate symptom inquiries into epidemic control protocols.
• For resource-limited settings, tracking symptomatic exposed individuals is a minimum effective strategy.